2010年4月28日

Increasing the number of open file descriptors – LINUX

Increasing the number of open file descriptors – LINUX, take 131072 as example.

1, increase this to 131072 for all users


vi /etc/security/limits.conf



* soft nofile 131072



* hard nofile 131072



2, ulimit –n 131072



3, You may have to modify ssh and login in pam.d. Because if not, you open a session using ssh, it will not take effect.



vi /etc/pam.d/sshd



Add



session required /lib/security/pam_limits.so



vi /etc/pam.d/login



Add



session required /lib/security/pam_limits.so



4, service sshd restart


Full Text

2010年4月21日

Replacing a failed drive in a Sun Server

Per our friend George Pagan we should be using the following procedure to replace failed/failing disks in Sun Servers:

Replacing failed sun disk

sudo vxdiskadm

Option 4 to remove the disk

Then from the OS

sudo cfgadm -f -y -c unconfigure controller::dsk/target

For example if c1t0d0 is the disk to be removed:

sudo cfgadm -f -y -c unconfigure c1::dsk/c1t0d0

This will turn on either an amber light on older machines or a blue light

on newer machines. Then it is safe to remove the drive.

After you insert the drive you run, for example:

sudo cfgadm -c configure c1::dsk/c1t0d0

Then do

devfsadm

vxdctl enable

And then option 5 in vxdiskadm to replace the disk.


Full Text

2010年4月16日

ERROR: V-3-21268: /dev/vx/dsk/dgD280silo1/orgvol is corrupted. needs checking

 

Sometime when you migrate from primary to secondary, the volume might be corrupted, just use fsck to fix it:

 

fsck -F vxfs /dev/vx/rdsk/dgD280silo1/orgvol
log replay in progress
replay complete - marking super-block as CLEAN
# mount -F vxfs /dev/vx/dsk/dgD280silo1/orgvol /orgvol


Full Text

2010年3月15日

recovery from RVG is disabled

    sometims your RVG cannot be started anyway, and the “vxprint –Vl” shows something like

    “kernel=DISABLED”

    In my case, the srl volume status is NEEDSYNC, that makes the RVG cannot be started, so repair the srl volume with following instructions:

1.  Dissociate SRL:

# vxvol -g vrdg -f dis vvrsrl

vxvm:vxvol: WARNING: Rvg rvg1 needs recovery. Volume being dissociated may not be up-to-date

2.  Stop the SRL:

# vxvol -g vrdg stop vvrsrl

3.  Start the SRL. It will take some time for the SRL to resynchronize.  You can run vxtask list from another window to monitor progress.

# vxvol start vvrsrl

4.  Re-associate the SRL:

# vxvol -g vrdg aslog rvg1 vvrsrl

5.  Recover the RVG:

# vxrecover -g vrdg -s

6.  Start the RVG:

# vxrvg start rvg1

Reference from symantec:

http://seer.entsupport.symantec.com/docs/268215.htm


Full Text

2010年3月8日

转-恢复unstartable的卷

root@com00biiacc002:~ #> vxprint -g COM2_DATA_DG -hvt
V  NAME         RVG/VSET/CO  KSTATE   STATE    LENGTH   READPOL   PREFPLEX UTYPE
PL NAME         VOLUME       KSTATE   STATE    LENGTH   LAYOUT    NCOL/WID MODE
SD NAME         PLEX         DISK     DISKOFFS LENGTH   [COL/]OFF DEVICE   MODE
SV NAME         PLEX         VOLNAME  NVOLLAYR LENGTH   [COL/]OFF AM/NM    MODE
SC NAME         PLEX         CACHE    DISKOFFS LENGTH   [COL/]OFF DEVICE   MODE
DC NAME         PARENTVOL    LOGVOL
SP NAME         SNAPVOL      DCO
EX NAME         ASSOC        VC                       PERMS    MODE     STATE

v  COM_DATA_2   -            ENABLED  ACTIVE   312475648 SELECT   -        fsgen
pl COM_DATA_2-01 COM_DATA_2  ENABLED  STALE    312475648 CONCAT   -        WO
sd COM2_DATA_DG02-01 COM_DATA_2-01 COM2_DATA_DG02 0 312475648 0   AMS_WMS1_6 RLOC
pl COM_DATA_2-02 COM_DATA_2  ENABLED  ACTIVE   312475648 CONCAT   -        RW
sd COM2_DATA_DG01-01 COM_DATA_2-02 COM2_DATA_DG01 0 312475648 0   AMS_WMS0_6 ENA
dc COM_DATA_2_dco COM_DATA_2 COM_DATA_2_dcl
v  COM_DATA_2_dcl -          DETACHED DETACH   21888    SELECT    -        gen
pl COM_DATA_2_dcl-01 COM_DATA_2_dcl DISABLED RECOVER 21888 CONCAT -        RW
sd COM2_DATA_DG02-02 COM_DATA_2_dcl-01 COM2_DATA_DG02 312475648 21888 0 AMS_WMS1_6 ENA

发现卷的状态变成detached,尝试start volumn

root@com00biiacc002:~ #> vxvol -g COM2_DATA_DG start COM_DATA_2_dcl
VxVM vxvol ERROR V-5-1-1198 Volume COM_DATA_2_dcl has no CLEAN or non-volatile ACTIVE plexes

start volumn不成功,因为不是clean或active的plex. 下面查看一下dg状态信息

root@com00biiacc002:~ #> vxinfo -g COM2_DATA_DG
COM_DATA_2     fsgen    Started
COM_DATA_2_dcl gen      Unstartable

发现COM_DATA_2_dcl 是unstartable的卷。尝试将COM_DATA_2_dcl的plex设置为clean状态。

root@com00biiacc002:~ #> vxmend -g COM2_DATA_DG fix clean COM_DATA_2_dcl-01
VxVM vxmend ERROR V-5-1-854 Plex COM_DATA_2_dcl-01 not in STALE state

根据提示将plex设置为stale状态。
root@com00biiacc002:~ #> vxmend -g COM2_DATA_DG fix stale COM_DATA_2_dcl-01
root@com00biiacc002:~ #> vxprint -g COM2_DATA_DG -hvt COM_DATA_2_dcl
V  NAME         RVG/VSET/CO  KSTATE   STATE    LENGTH   READPOL   PREFPLEX UTYPE
PL NAME         VOLUME       KSTATE   STATE    LENGTH   LAYOUT    NCOL/WID MODE
SD NAME         PLEX         DISK     DISKOFFS LENGTH   [COL/]OFF DEVICE   MODE
SV NAME         PLEX         VOLNAME  NVOLLAYR LENGTH   [COL/]OFF AM/NM    MODE
SC NAME         PLEX         CACHE    DISKOFFS LENGTH   [COL/]OFF DEVICE   MODE
DC NAME         PARENTVOL    LOGVOL
SP NAME         SNAPVOL      DCO
EX NAME         ASSOC        VC                       PERMS    MODE     STATE

v  COM_DATA_2_dcl -          DETACHED DETACH   21888    SELECT    -        gen
pl COM_DATA_2_dcl-01 COM_DATA_2_dcl DISABLED STALE 21888 CONCAT   -        RW
sd COM2_DATA_DG02-02 COM_DATA_2_dcl-01 COM2_DATA_DG02 312475648 21888 0 AMS_WMS1_6 ENA
root@com00biiacc002:~ #>

再次尝试start volumn.
root@com00biiacc002:~ #> vxvol -g COM2_DATA_DG start COM_DATA_2_dcl
VxVM vxvol ERROR V-5-1-1198 Volume COM_DATA_2_dcl has no CLEAN or non-volatile ACTIVE plexes

由于plex不是clean或active状态而无法start volumn. 将plex设置为clean状态。
root@com00biiacc002:~ #>
root@com00biiacc002:~ #> vxmend -g COM2_DATA_DG fix clean COM_DATA_2_dcl-01

start volumn.
root@com00biiacc002:~ #> vxvol -g COM2_DATA_DG start COM_DATA_2_dcl
root@com00biiacc002:~ #> vxprint -g COM2_DATA_DG -hvt COM_DATA_2_dcl
V  NAME         RVG/VSET/CO  KSTATE   STATE    LENGTH   READPOL   PREFPLEX UTYPE
PL NAME         VOLUME       KSTATE   STATE    LENGTH   LAYOUT    NCOL/WID MODE
SD NAME         PLEX         DISK     DISKOFFS LENGTH   [COL/]OFF DEVICE   MODE
SV NAME         PLEX         VOLNAME  NVOLLAYR LENGTH   [COL/]OFF AM/NM    MODE
SC NAME         PLEX         CACHE    DISKOFFS LENGTH   [COL/]OFF DEVICE   MODE
DC NAME         PARENTVOL    LOGVOL
SP NAME         SNAPVOL      DCO
EX NAME         ASSOC        VC                       PERMS    MODE     STATE

v  COM_DATA_2_dcl -          ENABLED  ACTIVE   21888    SELECT    -        gen
pl COM_DATA_2_dcl-01 COM_DATA_2_dcl ENABLED ACTIVE 21888 CONCAT   -        RW
sd COM2_DATA_DG02-02 COM_DATA_2_dcl-01 COM2_DATA_DG02 312475648 21888 0 AMS_WMS1_6 ENA

volumn及plex恢复正常。


Full Text